See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/323536714

Intelligent Transportation System (ITS) for Smart-Cities using Mamdani Fuzzy

Inference System

Article  in  International Journal of Advanced Computer Science and Applications · January 2018

DOI: 10.14569/IJACSA.2018.090215

CITATIONS READS
61 4,138

5 authors, including:

kashif iqbal Shah Muhammad Adnan Khan

Government College University, Lahore Gachon University
6 PUBLICATIONS   138 CITATIONS    239 PUBLICATIONS   4,155 CITATIONS   

SEE PROFILE SEE PROFILE

Sagheer Abbas Zahid Hasan

National College of Business Administration & Economics National College of Business Administration & Economics
133 PUBLICATIONS   2,756 CITATIONS    2 PUBLICATIONS   75 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Energy Demand Forecasting Using Fused Machine Learning Approaches View project

Deep Extreme Machine Learning View project

All content following this page was uploaded by Muhammad Adnan Khan on 05 December 2018.

The user has requested enhancement of the downloaded file.

 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 9, No. 2, 2018

Intelligent Transportation System (ITS) for

Smart-Cities using Mamdani Fuzzy Inference System
1 2 3
Kashif Iqbal Muhammad Adnan Khan, Areej Fatima
Government College University Sagheer Abbas, Zahid Hasan Department of Computer Science
Lahore, Pakistan School of Computer Science Lahore Garrison University,
National College of Business Lahore, Pakistan
Administration & Economics,
Lahore, Pakistan

Abstract—It is estimated that more than half of the world
population lives in cities according to (UN forecasts, 2014), so
cities are vital. Cities, as we all know facing with complex
challenges – for smart cities the outdated traditional planning of
transportation, environmental contamination, finance
management and security observations are not adequate. The
developing framework for smart-city requires sound
infrastructure, latest current technology adoption. Modern cities
are facing pressures associated with urbanization and
globalization to improve quality-of-life of their citizens. A
framework model that enables the integration of cloud-data,
social network (SN) services and smart sensors in the context of
smart cities is proposed. A service-oriented radical framework
enables the retrieval and analysis of big data sets stemming from
Social Networking (SN) sites and integrated smart sensors
collecting data streams for smart cities. Smart cities’
understanding is a broad concept transportation sector focused
in this article. Fuzzification is shown to be a capable
mathematical approach for modelling traffic and transportation
processes. To solve various traffic and transportation problems a
detailed analysis of fuzzy logic systems is developed. This paper
presents an analysis of the results achieved using Mamdani
Fuzzy Inference System to model complex traffic processes.
These results are verified using MATLAB simulation.
Keywords—Information Communication Technology (ICT);
Internet of Things (IoT); Intelligent Transportation System (ITS);
Fuzzy Inference System (MFIS); Traffic Congestion Conditions
(TCC); SNA; MF; Mamdani Fuzzy Inference System (MFIS)
I. INTRODUCTION
It is the time of Social Networking, Cloud Computing and
explosion of smart sensors deployed everywhere [1].
According to UN survey in 2014, more than half of world's
population now living in urban areas [2] and increasing surely
alerting city planners. Connected cities emerge when Internet
of Things (IoT) technologies and socially-aware network
systems aggregate administrations over a whole connected
metropolitan territory. When thinking of connected urban
areas, one may think of high tech cities that have the
prominent cutting-edge technologies for their citizens like
Copenhagen, London, New York, Chicago, Stockholm or
Amsterdam. However, small residential communities have
also been benefiting from interfacing individuals,
administrations, city infrastructure and services. This article
investigates city transportation problem and a portion of the
difficulties that are involved with developing widespread IoT
techniques. The coalition of world-class IoT improvement
anticipates working with each of these smart urban
communities that enable citizens to make technology
utilization more sensible, adaptable and sustainable. Many
urban cities and towns around the globe are turning to socially
connected smart devices to solve urban problems [3], for
example, traffic congestion, environmental contamination,
healthcare, security surveillance to enhance the living
standards for their general public everyday comforts. Smart
sensors that are installed throughout the city, in vehicles, in
buildings, in roadways, in control monitoring systems,
security surveillance and applications and devices that are
utilised by individuals who are living or working in the city
[4]. Delivering information to the public that is utilizing
through these high tech smart cities opportunities. The big-
data analytics utilized to decide on how public spaces are
planned, how to make the best utilization of their assets and
how to convey administrative notifications more proficiently,
viable and appropriately [5].
Therefore, most urban cities have embraced huge
investments during recent decades in Information
Communication Technology (ICT) infrastructure including
computers, broadband availability and some sensing
frameworks [6]. These infrastructures have engaged various
inventive administrations in territories, for example,
demographic sensing, urban coordination and real information
that makes living ones close. Such administrations have been
widely sent in a few urban cities, accordingly exhibiting the
potential advantages of ICT frameworks for organisations and
the natives themselves [7]. During most recent years it has
additionally seen a blast of sensor distribution, along with
the development of adaptive systems, internet‐of‐things [8]
current advancements of sensor-based systems have emerged.
Currently, the advantages of social communication and
internet‐of‐things distributions for smart urban areas have
likewise been exhibited [9].
Current Smart City data analysis implies complex stream
analytics for a comprehensive set of activities aiming to turn
into real actionable outcomes [10]. The analysis comprises of
following contributions:

94 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
1) Analysis of thousands of traffic blockage cases, road
capacity measures, traffic signalling and dynamic, consistent
information to give a better message to the citizens.
2) Events, episodic road examination, utilising real-data
gathered by citizens, devices and sensors.
3) Turning web-based into social media information,
important city events analysis, assumptions, examinations,
and numerous other things. Consolidating information from
physical (sensors/devices) and social sources (social
organisations) can give full, essential information and adds to
better assessment and bits of knowledge.
Over-all speaking, smart cities realization is a broad
concept so, the transportation sector is focused in this article.
Fuzzy logic is one of the strongest candidate solution for
mathematical based modelling. In this article fuzzy logic-
based solution is proposed for transportation problem. The
input parameters are: Vehicular Speed (VS), Road Capacity
(RC), Traffic Signals (TS), Trip Riding Distance (RD) and
Distance Traffic Signals (DTS). A detailed Transportation
fuzzy logic system is developed based on rule-based
inferencing to solve the traffic congestion issues. Analysis of
the results obtained using Mamdani Fuzzy Inference System is
verified using MATLAB Simulation
The objective of this paper is to analysis key issues and the
solutions about traffic congestion in a smart city in the light of
critical inducing aspects. The rest of the paper is structured as
follows: Section II gives an overview of related and similar
works that can be found in the international literature. Section
III presents the fundamental architecture and approach.
Sections IV and V presents technical details, sentiment
analysis of problems and a conceptual model for smart-cities.
Section VI provides a proposed Mamdani Fuzzy Inference
System (MFIS) based results analyses, the work is planned in
the context of simulation and Section VII contains conclusion
and future work to be planned in the context of smart drive
mobile apps.
II. LITERATURE REVIEW
A smart IoT system which automatically notifies necessary
information of passengers after triggering of shock detector
sensors to lowering loss rates in accidents and alert nearby
local public safety organization about the physical location of
accident suggested by Nasr et al. [11]. Rizwan et al.
industrialize a smart traffic management system roadside unit.
It carries alternate routing to avoid traffic blocking and
increase traffic flow through IoT and lower traffic density,
offers predictive analytic technique (Big-data techniques)
[12]. Scalable Enhanced Road Side Unit, SERSU, proposed
by Al-Dweik et al. used wireless communication network and
radio frequency adaptive traffic control system, pollution
detection system and weather information system. SERSU
components were placed on the roadsides with various breaks,
capturing generated sensor signals by vehicle sensors module
[13]. Modern techniques in cars, internet and their current and
future relationship, detail history of usage of electronic
devices in automobiles, and social implication of these
technologies briefly studied by Goggin [14]. Joshi et al. made
infrared-based sensor system, which to monitor traffic flow
www.ijacsa.thesai.org
Vol. 9, No. 2, 2018
and provides alternate road traffic routing path to drivers for
the avoidance of traffic crowding capture infrared radiations
emitted by vehicles on road surface [15]. Handte et al.
designed IoT enabled the navigational system for real
transport facility, provided complete guidance of routes to bus
riding passengers for urban bus riders in Madrid, which were
assisting in micro-navigation, expects massive aware routes. A
system to communicate with onboard sensors to sense the
presence of onboard passengers, this system was based on
mobile devices. Their system collected real-world bus user’s
response for better accessibility of travel information [16].
Zanella et al. advised web-based service approach for IoT
service architecture to resolve integration issues for different
end node devices connected to IoT system Zanella et al. also
evaluated key ideas, facilities and solution are currently
available for implementation of IoT based smart cities [4].
Technological challenges and socio-economic opportunities in
developing and designing of future smart cites discussed key
by Theodoridis et al., they also suggested 3-tier IoT nodes and
3-plane architecture model. Further, they develop a city scale
test bed for future internet and IoT experimentation [17]. A
hierarchy which combines smart homes and smart cities
described by Skouby et al., they also proposed a four-layered
model to join end nodes IoT devices, communication
technologies like distributed artificial intelligence and cloud of
things [6]. Gubbi et al. presented Radio Frequency
Identification (RFID’s) a user-centric cloud-based vision of
implementation of IoT, by the interaction of public and private
clouds, major research trends, IoT application domain, current
and future enabling technologies etc. that will drive IoT
shortly [14]. Base Station arrangement, based architecture
sensor system for intelligent traffic light system (TLS)
suggested by Chong et al. They designed intelligent software,
implemented on TLS which continuously communicates with
the base station and calculates green light time, and provide
monitoring of traffic by officers [18].
Internet of Vehicle (IoV), a unique solution for smart
traffic management is discussed by Dandala et al. They argued
that IoV can be an effective solution conventional IoT based
traffic management technique to overcome traditional traffic
issues. Further, they described to be a reality which is a
vehicle to vehicle’s owner that IoV needs four types of
communication, a vehicle to vehicle, a vehicle to centralize
server and vehicle to the third party like police patrol,
ambulance, etc. [19]. Cognition was used for user
authentication in vehicles [26]. Sagheer et al. proposed a
fuzzy inference system to avoid traffic congestion using bio-
inspired method [27].
Density-based signalling to overwhelmed issues raised by
fixed time signalling for example in fixed time signalling
method the traffic lights have predefined periodic time system
suggested by Thakur et al. provides intelligent signalling by
assigning the greener signal to dense traffic region to avoid
congestion by continuously evaluating traffic density [20].
Ramchandra et al. proposed a comparable system which
device traffic lights by using average speed of vehicles
dynamically according to the density of traffic. In this
proposed system every vehicle is equipped with On-Board
95 | P a g e
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Device (OBD) distribute data to centralise server using Zigbee
protocol which acquires vehicle speed data process [21].
Chowdhury et al. proposed intelligent traffic light system
for messaging between emergency vehicles infrastructure and
to reduce traffic congestion and increase reliability to traffic
signals. The proposed system considers the priority of vehicle
depends on the type of incident and to secure signals from
hacking [22]. Some shortcomings in the traditional intelligent
transportation system and argued to prefer Radio Frequency
Identification (RFID) pointed out by Ou et al., sensor system
and networking technologies to overwhelmed traditional
intelligent transportation systems [23].
Information-Centric Networking to project and device
Future Internet Architecture proposed by Amadeo et al. In
Information-Centric Networking which uses IoT submissions
to access data of every end node device having unique
location name [24].
III. SMART-CITY KEY FEATURES REALIZATION
The availability of smart solutions for cities has risen
quickly over the most recent years. Therefore, technical
solutions exist for each city to become smarter. The challenge
today is mostly to execute proper solutions proficiently, as
opposed to just concentrating on innovations. Smart city areas
cannot be developed through a patchwork approach, yet by the
well-ordered adoption of incremental changes. The most
proper way of smart-city realisation is introducing a smart
system working group of volunteers characterize its
manageability vision and afterwards lays out an electronic
well-ordered guide and execution design. The capacity to
distinguish the acutest bottlenecks to send coordinated and
flexible solutions and afterwards to use these outcomes into
other smart community’s activities requires involvement and
strong specialised expertise.
Smart City Key Resources: Transportation, Climate
Change, Energy, Utilities, Security Surveillance, Healthcare,
Business Management, etc.
Connected cities enhance the experience of workers by
analyzing data and smart city coordinators by breaking down
information from reporting frameworks including sensors,
roadside cameras, brilliant monitoring systems and speed
check signs. Applying IoT innovations to solve urban
community’s issues includes gathering the information that is
collecting from sensors, recordings by cameras, interpersonal
organizations and brilliant devices that are examining real-
data. This data is delivered noteworthy bits of knowledge that
are utilized straightforwardly to trigger actuators that are
associated with smart devices. For example, versatile smart
city assets, connected by implications, to illuminate choices on
policy and to streamline jobs. In smart urban communities,
these arrangements include monitoring geographic
information from Global Positing System (GPS) trackers and
RFID labels on vehicles [15], structures, buildings and power
stations, breaking down the proceed of vehicles to recognize
occurrences or blockages. Smart buildings security,
interpersonal organizations, city administrations are
straightforwardly modifying frameworks continuously to
control the activity stream in city events, security observation
www.ijacsa.thesai.org
Vol. 9, No. 2, 2018
investigation and reduce traffic delays. Authentic analysis of
city traffic, security investigation and movement blockage and
roadside sensors information can likewise be utilized to alter
time delays, misinterpreting security observation, speed cutoff
points and city toll tax, control security monitoring and
activity stream in the more flows for long-term outcomes. To
route movement around incidents, sensors additionally write
about the state of streets conditions, weather updates,
buildings structures, road lights and extensions with the goal
that support to schedule maintenance when required.
Smart cities will make emerging activities in
transportation, utilities, smart buildings and smart security.
Smart city design plan leaders shaped a working group of
ecosystem system accomplices to evaluate robust city
community’s abilities and guide a long-term vision that
coordinates with the city's future planning. Smart city planners
have endorsed digitalising citywide assets like fast travel
framework, smart buildings, smart security, electric transport
and is additionally pushing ahead for far-reaching IoT hub that
will pioneer digital city infrastructure.
Designing a roadmap for smart-cities is based on four core
pillars: Connectivity is the foundational layer of a smart-city.
In real-time data is collected about peoples, places and things
by smart sensors and this data are stored on cloud application
servers to analyze and utilised to take better real-time
decisions and planning as shown in Fig. 1.
Mobility means moving peoples, goods and information
efficiently and efficiently. The economic-mobility means
regardless of circumstances online job seekers in smart-cities
find maximum jobs available that are not handy via public
transportation.
Next is security improving public and private places
security, data protection and cyber-security while using latest
ICT’s technologies on-line and off-line.
Sustainability, of course, is focusing on sustainable
practices in critical sectors of cities such as transportation,
energy consumption, climate change, utilities, security
observations, and financial services.
Implementation of smart-cities solutions may have three
things every day for their citizen, i.e., creates values, generates
revenues and cut costs depending on value exchange smart
systems and smart projects.
Fig. 1. Road-map four pillars of a smart-city.
96 | P a g e
 (IJACSA) International Journal of Advanced Computer Science and Applications,
A. Smart Cities Framework Model Overview
The adaptive data analysis stage is plotted out in the
background. It is made out of different layers, bring down a
level (devices, communication planes), middle layers (data,
information examination) at higher layers (application,
dashboard planes). At each layer, distinctive programming
code chunks perform specific operations, related to collecting
data, messaging, data accessing, semantic annotation,
examination or perception where applications can join
segments from different layers in light of their specific pre-
requisites. Along these advances toward getting to be plug and
play and can be mainly used in smart-city sectors applications.
The present extraordinary cutting-edge advances of portability
smart-phones, interpersonal organization services and objects
are coordinated together for a new time machine to machine
and person to person communication correspondence [17].
1) Main Components of the Model
The layered framework model of a smart city as shown in
Fig. 2 is having four main layers described below as:
a) Sensing Layer
Sensor Layer comprises tens to thousands of sensor hubs
connected using smart remote technologies. They gather data
from the environment and convey it to other connected
devices that pass the data to the cloud server over the Internet.
b) Communication Layer
Wireless heart innovative technology gives excellent
remote protocol access to the full range of processors capacity,
control, and resource management to applications. DigiMesh
is an exclusive shared systems networking topology for use in
remote end-point network connectivity through the physical
Internet.
c) Data Layer
The capacity and processing of data should be possible on
the edge of the networks itself or in a cloud server. If any
preprocessing of data is a need, then it is typically done at
either the sensor or some other proximate device.
Fig. 2. Layered Model of a smart-city.
www.ijacsa.thesai.org
Vol. 9, No. 2, 2018
The processed data is then regularly sent to a remote
server. The capacity and processing abilities of an IoT object
are additionally controlled by the assets accessible, which are
regularly exceptionally compelled because of constraints of
capacity, vitality, control, and computational ability.
d) Application Layer
The application layer is responsible for data organization
and presentation. The application layer on the Internet is
regularly in light of Hyper Text Transfer Protocol (HTTP)/
File Transfer Protocol (FTP) standards. The proposed events
in this examination are sharing of dynamic data to customers
using mobile phones as a particular device. It may be HTTP is
not reasonable in resource enabling situations since it is
relatively verbose and this manner brings about a significant
parsing overhead. Many other innovative conventions have
been produced for IoT resources, for example, Message Queue
Telemetry Transport (MQTT) and Constrained Application
Protocol (CoAP).
Along with these four layers following components move
toward becoming plug and play real-time integration and
stream-analytics that can be utilized explicitly by specific
smart applications framework by adaption of these technical
modules given below:
a) Data Wrapper
It is a program that extracts the content from a particular
information source and translates it into an organization
format. Using sensory meta-data, it extends a generic way to
describe features of sensors, about the data stream that
containing general information. A semantic annotation module
annotates the sensory parsed data.
b) Data Aggregation
For data aggregation, the source information originates
from public records online databases. The information is
packaged into aggregate reports this information is useful for
business, marketing, local and government organizations. It
reduces the large volume of data, i.e. the size of raw sensory
observations delivered by the data wrappers by using data
compression techniques and time series analysis.
c) Data Federation
Answers to user queries, according to the requirements it
first finds the relevant stream. It then translates the user
request into Resource Description Framework-Stream
Processing (RDF-RSP) queries and obtains results
accordingly. As fast changing real-world data from sensors
and online services evolves IoT-based smart environment
monitoring, real-time processing and analytics based on RSP
semantics. RDF query language manages continuous data
streams SPARQL, and CQELS languages support RDF
reasoning.
d) Event Detection
The event detection is the identification of items, events
and observations, i.e. constraints on what defines an event is
relaxed or usually modelled as a set of thresholds or
probabilities. In city sectors, it provides tools or web
software’s applications that monitor urban areas events such
as the need for clearing transport deadlock, emergency
97 | P a g e
 (IJACSA) International Journal of Advanced Computer Science and Applications,
facilities, irrigation facilities, pest identification in crops and
growing view of the traffic congestion form real-time
annotated and aggregated data streams.
IV. CONCEPTUAL FRAMEWORK FOR SMART-CITIES
The proposed technique in this study is sharing of dynamic
information to users using smartphones as a communication
device. Current day advanced modern technologies of
knowledge mobilization, cloud-servers and the smart-city app
are integrated into a new era of quick communication.
Advanced smart-phones may have limited sensing capabilities
but enhanced computational strength, lesser cost, excessive
usage, availability of Global System for Mobile
communication (GSM) and mobile internet signals,
availability of different sensors in smartphones like a
gyroscope, digital compass, proximity sensor, etc. Services
available like Google map, Google weather, IBM live
streaming analytics etc. is prime motivation to use a
smartphone as sensor I/O device in the proposed system.
Moreover, specialised and more accurate sensors like
accelerometer, Global Positioning System (GPS), and shock
sensor etc. services are realised. Also, specific and more
accurate sensors like accelerometer, global positioning system,
and amaze sensor so on so forth are outlined and created on
various stages and new technologies integrating with existing
technologies in a single integrated system that is beyond the
scope of the current proposes a study as shown in Fig. 3.
The proposed system aims to provide efficient and
effective smart cities traffic infrastructure. In this study, we
show the concepts of cloud computing, bid-data analysis,
internet of things, human-computer interactions, software
engineering paradigm etc. can be the realization of smart-
cities traffic framework to improve the living standards of
their citizens.
Fig. 3. Proposed Model of Smart City App.
www.ijacsa.thesai.org
Vol. 9, No. 2, 2018
Studies suggest that smart cities need specific information
for experiencing globalization by making efficient smart city
decisions like smart transportation, smart energy distribution,
demo-graphic information, smart utilities, healthcare
services, etc.
Travelers from one city to another city have very little
information about nearby pinpoint spatial locations, safety
organisations, emergency services and government building
and necessary information for traveller and visitors with them.
In case of any emergency situation, even local public safety
organizations have no personal, medical records or emergency
contact numbers for any situation. On any highways, peoples
hesitate to do or accept any help from others travellers or
unknown peoples because of no information. To overcome
these issues with smart sensors, surveillance cameras,
Wireless Fidelity (Wi-Fi) devices may help the citizen of
smart-cities by the following critical technical innovations:
 Users will get real-time and dynamic information about
the city routes with other cities in a particular range by
Google Map as shown in Fig. 4. Smart mobile-app
collected device and personal mobile information for
notification, newsfeeds in text or audio format.
 Smart city will alert their citizens about road
congestion in the form of text, audio/video format.
 Provide necessary information of other cities around
and also provides a platform to communicate with
these connected cities by text messages.
 Track record of smart cities user is or travellers from
start to destination and generates alerts of essential
places nearby like fuel stations, restaurants, hospitals,
emergency stations, etc.
 Smart cities are real-data streaming analytics which
provides complete details of their citizen with utility
services to transportations facilities.
 Users can send and receive with one push any
emergency messages to other cities users (using
GSM/GPS, other Internet services) as well as inform
nearby emergency organizations in the form of an
email as shown in Fig. 5.
 Users can comfortably offer or accept emergency
pickup, health-care services, nearby building info and
share visits from other travellers especially on
highways because smart cities will keep track of these
connected cities.
 Reporting of any crime, security surveillance, weather
forecast, misconduct to authorities nearby (if
witnessed) with proper privacy.
 The user receives text messages as well as audio/videos
format to prevent mental divergence.
98 | P a g e
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Fig. 4. Sample prototype of Smart-Map.
Fig. 5. Prototype of Smart-City App.
Smart cities users are using user’s smartphone as sensing
as well as a communication device. In the proposed system
smartphones will act as wireless sensor network’s node.
Internet and Mobile telecommunication GSM signals will act
a medium of communication between all wireless networked
sensor nodes. The application server will host smart-cities
application and is connected to Short Message Service (SMS)
server which will generate text messages, and it is also an
interface between end nodes and application cloud-server. The
application server will also send e-mails to public safety
organizations in case of an emergency. The application server
www.ijacsa.thesai.org
Vol. 9, No. 2, 2018
will provide all the necessary computations. Microsoft Azure
IoT cloud server will be used because of its enhanced features
for smart cities utility services, transportation conditions,
environmental conditions and security features realization.
V. SOCIAL ANALYSES
In this section social analysis of smart cities as a sample
has been performed in the form of graphical representation. A
graph is a data structure which consisting of a finite number of
edges and nodes. There are many ways to represents a node,
edge graph, for example, adjacency matrix, graph ML format,
CSV files.
The adjacency matrix is a two-dimensional square matrix
whose size is equal to the number of nodes in the graph.
However, if input graph contains a large number of nodes and
less number of edges then the adjacency matrix became sparse
and space consuming. Fig. 6 represents a sample connected
cities graph and Table I represents an adjacency matrix 6*6 of
connected cities.
In the graph illustrated in Fig. 6 nodes represent cities at
different ranges and edges for instance roads between cities,
paths and connectivity or relationship between cities are in
different ranges.
Fig. 6. Connected smart cities graph.
Edges define the relationship between different users or
cities resources, a directed edge from city 1 to city 2
represents that city one can communicate with city 2 and
city 3. City 2 can communicate with city 5, city 4 and city 3
and so on for every city connection in a graphical format
having close centrality measures. The adjacency matrix is
represented in Table I in this “0” represents no relationship,
and “1” represents the positive relationship. If city one users
want to communicate with city four user’s, they can
communicate with the help of city two based on shortest path
algorithm between two nodes traversed.
99 | P a g e
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 9, No. 2, 2018

TABLE I. ADJACENCY MATRIX BETWEEN CITIES 6 X 6 VI. PROPOSED MFIS BASED SOLUTION
This section explains in detail Mamdani Fuzzy Inference
City 1 City 2 City 3 City 4 City 5 City 6 System (MFIS) based on smart-city Traffic Congestion
Conditions (TCC) controls. The facts given below explain the
City 1 0 1 1 0 0 0 measuring of TCC for the Smart-city for smart drive facility
which is based on Mamdani Fuzzy logic principles.
City 2 1 0 1 1 1 1 In this article, the planned MFIS which is capable of
measuring the TCC for the city algorithm is given in Table III.
City 3 1 0 0 1 0 0 The five inputs and one output MFIS is proposed to calculate
TCC.
City 4 0 1 1 0 0 0
In this method five inputs that are: Vehicle Speed (S),
City 5 0 1 0 0 0 1 Load Capacity (C), Traffic Signals (T), Distance between
Signal (D), Riding Distance (R) are taken. These inputs are
City 6 0 1 0 0 1 0 used to build up a lookup table given in Table IV to decide
TCC for a respective algorithm for input-output relation given
This phenomenon is used on the higher level as well in by MFIS. Its mathematical representation is shown in (1).
computational intelligence. In which every node represents a µCG = MFIS [µVS, µRD, µTS, µDTS, µRD] (1)
smart city a cluster of cities and edges represents any one of
cites which can reside or act as the interaction between two In this article, the Intelligent Transportation System (ITS)
groups or clusters. The central city would be helpful for is measured using Mamdani Fuzzy Inference System (MFIS).
communication between cities in different geographical Table I shows the proposed MFIS Based ITS algorithm. The
location city areas might be other cities. This technique will I/O surface for MFIS is given in Fig. 1.
enhance the range of communication between two distanced
cities. The model of communication of distant (out of the TABLE III. PROPOSED MFIS BASED TCC ALGORITHM
range) cites clusters shown in Fig. 7.
1. Inputs:
The adjacency matrix of connected cities is shown in In this system 5, Fuzzy Input variables are used which are the
Table II. following (Vehicle Speed, Road Capacity, Traffic Signal, Distance
Traffic Signal and Riding Distance)

2. Each Fuzzy Input variable has different types of membership

functions.

3. Every Fuzzy membership function is used to build fuzzy inference

rules.

Fig. 7. Communication of cities clusters.

TABLE II. ADJACENCY MATRICES OF DISTANT CITIES

Inter InterCity
Cluster 1 Cluster 2
City1 2

Cluster 1 0 1 0 0

Intermediate
1 0 1 0
City1

Intermediate
0 1 0 1
City2

Cluster 2 0 0 1 0
Fig. 8. Input and output surface for MFIS.

100 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 9, No. 2, 2018

TABLE IV. MATHEMATICAL AND GRAPHICAL MF OF MFIS INPUT VARIABLES

Sr. Input Membership Function(MF) Graphical Representation of MF

No.
[ ]
( ) {
µvs, slow [ ]}

1 VS [ ]
(µVS(S)) µvs, medium ( ) { }
[ ]

[ ]
µvs, fast ( ) { }
[ ]

[ ]
µrc,,narrow ( ) { [ ]}
RC (µRC(C))
2 [ ]
µrc, average ( ) { }
[ ]

[ ]
µrc,wide ( ) { }
[ ]

[ ]
µTS, fewer ( ) { }
[ ]

[ ]
µTS, average ( ) { }
[ ]
3 TS (µTS(T))
[ ]
µTS, much ( ) { }
[ ]

[ ]
( ) { ]}
µDTS, short
[

[ ]
µDTS, average ( ) { }
[ ]
DTS
4
(µDTS(D))
[ ]
µDTS, far ( ) { }
[ ]

[ ]
µRD, short ( ) { }
[ ]

[ ]
µRD, average ( ) { }
RD [ ]
5
(µRD(R))
[ ]
µRD, long ( ) { }
[ ]

101 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 9, No. 2, 2018

about 81 input-output rules, the system complexity increased

TABLE V. INPUT VARIABLE RANGES if the number of rules increased. The Mamdani Fuzzy
Inference rules are shown in Fig. 9.
Input
Sr # Ranges Semantic sign
Parameters E. Inference Engine
0-50 Slow The Mamdani Inference Engine is used to map five inputs
1 VS 45-75 Medium
to one output (TCC) as shown in Fig. 8.
60-120 Fast
0-50 Narrow
2 Cap 40-100 Average
90-150 Wide
0-5 Fewer
3 TS 4-10 Normal
9-15 Too much
2-4 Nearer
4 DTS 3-6 Average
5-10 Far
0-10 Nearer
5 RD 8-20 Center
18-30 Far

A. Input Fuzzy Sets

Fuzzy input variable is statistical values that are used to
calculate the Traffic Congestion Condition in smart cities.
In this article, five different types of fuzzy variables are
used for the analysis of congestion in smart cities. The detail
Fig. 9. I/O Rules for ITS.
of these input variables is given in Table V.
B. Fuzzy Output Variable F. De- Fuzzifier
Fuzzy output variable Traffic Congestion Control (TCC) is In this article centroid, De- Fuzzifier is used. Fig. 10 to 12
used to calculate the result by the values of input variables in represents rule surface of Proposed ITS using MFIS.
the world of discourse. The details of output are shown in Fig. 10 shows that if Vehicle Speed is between 1-80 km/s
Table VI. and Traffic Signals are lies in the range of 10 to 15, then
Traffic Congestion is approximately 80%, which is high.it also
TABLE VI. OUTPUT VARIABLE RANGES
shown that, if Vehicular Speed between 80 – 120 km/s and
Semantic sign for Traffic Signal is 10-15, then Congestion is low approximately
Sr # Output of MFIS Ranges 10%.
Congestion
0 - 0.5 No delay (Less)
Average delay
1 Congestion Control 0.2 - 0.7
(Medium)
0.5 - 1 Much delay (High)

C. Membership Functions
Membership function gives curve value between 0 and 1,
and it provides a mathematical function which provides
statistical values of input and output variable. Membership
functions are also available in MATLAB tool. The propose
solution uses the membership function which is as follows:
 Trim
Trim is a triangular curve built-in MATLAB function. To
calculation of this function, three scalar parameters are used in
the proposed solution which is Low, Medium, and High. The
mathematical equations and graphical representation of Fig. 10. Rule surface for traffic signals and vehicle speed.
membership function are given in Table IV.
Fig. 11 shows Congestion Control using input variables
D. Rule-Based Traffic signals and Riding Distance. It is observed that
In this system most, suitable rules for system congestion is approximately 80% when Traffic signals are 10-
understanding are applied. This rule-base system contains 15 and Riding Distance between Source to Destination lies in

102 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 9, No. 2, 2018

the range of 8 to 20 km. Congestion is approximately 60% if an efficient tool for programming, data analysis, visualisation
Traffic signals are 9-10 and Riding Distance grater then 8 km. and computing. For simulation results, five inputs and one
output Congestion Control variable is used.

TABLE VII. CONGESTION CONTROL BASED ON RULES DEFINED

Input Variables Output

VS RC TS DTS RD Congestion
H W L H H Less Delay
H N H L L Medium Delay
L N H L L High Delay

Table VII explains the rules of Proposed Congestion

Fig. 11. Rule surface for traffic signals and riding distance.
Fig. 12 shows Congestion depends upon input variables
Road Capacity and Riding Distance between source to
destination. Approximately, there is no Congestion when road
capacity is extensive (110 to 120 vehicles on the road) and
Riding Distance is 9 to 20 km. If Road Capacity between 100
to 110 approx and Riding Distance between 10 to 20 km then
congestion is increased upto 20% increase. If road capacity is
less than 90 (narrow road), then the congestion is up to 50%.
So, it concludes that Congestion inversely proportional to
Road Capacity.
Control system. Fig. 13 to 15 shows the proposed system
evaluation.
Fig. 13 shows the congestion is less if vehicular speed is
high and road capacity is wide. It further depicts that if a
traffic signal is few and distance between signals is high and
riding distance is far congestion is low.
Fig. 14 shows that congestion is Medium if the vehicle
speed is high and Road Capacity is Narrow, and traffic signals
are too much, and the distance between signals is low, and
riding distance is far then congestion is medium.
Fig. 15 explains congestion is high if the vehicle speed is
Low and Road Capacity is Narrow, and traffic signals are too
much, and the distance between signals is small, and riding
distance is also low than congestion is high.

Fig. 12. Rule surface for riding distance and road capacity.

G. Simulation Results
For simulation results, MATLAB R2017a tool is used.
Fig. 13. Lookup diagram for low TCC.
MATLAB is also used for modelling, simulation, algorithm
development, prototyping and many other fields. MATLAB is

103 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 9, No. 2, 2018

VII. CONCLUSION AND FUTURE WORK

Implementation of smart-cities infrastructure and services
is a long-game process [25]. The advantages of smart-city to
communities will not likely be quick and will probably be
incremental in the first step. Nevertheless, to accomplish
smart-cities infrastructure through the utilization of SNA,
Information Communication Technologies and IoT’s to scale
their city framework and extend services reasonably while
offering substantial financial advantages. This TCC fuzzy
expert system is designed with the help of 5 input and one
output variable. Mamdani Fuzzy Inference System (MFIS) is
used to evaluate the Traffic Congestion Conditions (TCC) in
smart-city. The proposed system design of TCC has been
beneficial to determine the city traffic congestion. Through
this system, anybody can check any traffic congestion. In
future, MFIS would be used to evaluate the performance of
the other resources of smart-city like Environmental
Conditions, Energy Consumption, Healthcare and Security
Surveillance, etc.
REFERENCES
1. Han, Q., Liang, S., & Zhang, H. (2015). Mobile cloud sensing, big data,
and 5g networks make an intelligent and smart world. IEEE
Network, 29(2), 40-45.
2. https://news.un.org/en/story/2014/07/472752-more-half-worlds-
population-now-living-urban-areas-un-survey-finds.
3. Carmona, Matthew. Public places, urban spaces: the dimensions of
urban design. Routledge, 2010.
4. Kitchin, Rob. "Big Data, new epistemologies and paradigm shifts." Big
Fig. 14. Lookup diagram for medium TCC. Data & Society 1.1 (2014): 2053951714528481.
5. Misuraca, Gianluca, Francesco Mureddu, and David Osimo. "Policy-
making 2.0: Unleashing the power of big data for public
governance." Open government. Springer, New York, NY, 2014. 171-
188.
6. Albino, Vito, Umberto Berardi, and Rosa Maria Dangelico. "Smart
cities: Definitions, dimensions, performance, and initiatives." Journal of
Urban Technology 22.1 (2015): 3-21.
7. Gabri Malek, Chunlin LI, Z. Yang, NajiHasan.A.H and X.Zhang,’
Improved the Energy of Ad hoc On-Demand Distance Vector Routing
Protocol’, International Conference on Future Computer Supported
Education, Published by Elsevier, IERI, pp. 355-361,2012.
8. Lakshmi, I. "A literature survey on Big Data Analytics in Service
Industry." International Journal of Engineering and Computer
Science 5.4 (2016). Breslin, J., Decker, S., 2007.
9. Dimitrakopoulos, George, and George Bravos. Current Technologies in
Vehicular Communication. Springer, 2016.
10. Psomakelis12, Evangelos, et al. "BIG IOT AND SOCIAL
NETWORKING DATA FOR SMART CITIES."
11. Nasr, Elie, Elie Kfoury, and David Khoury. "An IoT approach to vehicle
accident detection, reporting, and navigation." In Multidisciplinary
Conference on Engineering Technology (IMCET), IEEE International,
pp. 231-236. IEEE, 2016
12. Rizwan, Patan, K. Suresh, and M. Rajasekhara Babu. "Real-time smart
traffic management system for smart cities by using the Internet of
Things and big data." In Emerging Technological Trends (ICETT),
International Conference on, pp. 1-7. IEEE, 2016.
13. Al-Dweik, Arafat, Radu Muresan, Matthew Mayhew, and Mark
Lieberman. "IoT-based multifunctional Scalable real-time Enhanced
Road Side Unit for Intelligent Transportation Systems." In Electrical and
Computer Engineering (CCECE), 2017 IEEE 30th Canadian Conference
on, pp. 1-6. IEEE, 2017.
14. Goggin, Gerard. "Driving the internet: mobile internets, cars, and the
social." Future Internet 4, no. 1 (2012): 306-321.
Fig. 15. Lookup diagram for high TCC. 15. Joshi, Yashashree, Ashwini Joshi, Neha Tayade, Priyanka Shinde, and
S. M. Rokade. "IoT Based Smart Traffic Density Alarming Indicator."

104 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
(2016).Handte, Marcus, Stefan Foell, Stephan Wagner, Gerd Kortuem,
and Pedro José Marrón. "An Internet-of-Things Enabled Connected
Navigation System for Urban Bus Riders." IEEE Internet of things
journal 3, no. 5 (2016): 735-744
16. Theodoridis, Evangelos, Georgios Mylonas, and Ioannis
Chatzigiannakis. "Developing an IoT smart city framework."
In Information, intelligence, systems and applications (Lisa), 2013
fourth international conference on, pp. 1-6. IEEE, 2013.
17. Chong, Hon Fong, and Danny Wee Kiat Ng. "Development of IoT
device for traffic management system." In Research and
Development (SCOReD), 2016 IEEE Student Conference on, pp. 1-6.
IEEE, 2016
18. Chong, Hon Fong, and Danny Wee Kiat Ng. "Development of IoT
device for traffic management system." In Research and
Development (SCOReD), 2016 IEEE Student Conference on, pp. 1-6.
IEEE, 2016
19. Thakur, Tanvi Tushar, Ameya Naik, Sheetal Vasari, and Manjiri Gogate.
"Real-time traffic management using the Internet of Things."
In Communication and Signal Processing (ICCSP), 2016 International
Conference on, pp. 1950-1953. IEEE, 2016
20. Ramachandra, Sujit H., K. Nitesh Reddy, Vivek R. Vellore, Sumanth
Karanth, and Tareesh Kamath. "A novel dynamic traffic management
system using on-board diagnostics and Zigbee protocol."
www.ijacsa.thesai.org
View publication stats
Vol. 9, No. 2, 2018
21. In Communication and Electronics Systems (ICCES), International
Conference on, pp. 1-6. IEEE, 2016
22. Chowdhury, Abdullahi. "Priority-based and secured traffic management
system for an emergency vehicle using IoT." In Engineering & MIS
(ICE), International Conference on, pp. 1-6. IEEE, 2016
23. Ou, Haoyuan, Jianming Zhang, and Yi Wang. "Development of
intelligent traffic control system based on internet of things and fpga
technology in proteus." traffic 20 (2016): 2.
24. Amadeo, Marica, Claudia Campolo, Jose Quevedo, Daniel Corujo,
Antonella Molinaro, Antonio Iera, Rui L. Aguiar, and Athanasios V.
Vasilakos. "Information-centric networking for the internet of things:
challenges and opportunities." IEEE Network 30, no. 2 (2016): 92-100.
25. Coutard, Olivier, et al. "Urban Megatrends: Towards a European
research agenda." (2014): 1-17.
26. Saeed, Yousaf, et al. "Impact of Cognition on User Authentication
Scheme in Vehicle using Fuzzy Logic and Artificial Neural
Network." International Journal of Computer Science and Information
Security 14.10 (2016): 285.
27. Abbas, Sagheer, et al. "Bio-inspired neuro-fuzzy based dynamic route
selection to avoid traffic congestion." International Journal of Scientific
and Engineering Research2.6 (2011): 284-289.
105 | P a g e